Benchmarking TensorFlow on Cloud CPUs: Cheaper Deep Learning than Cloud GPUs

I've been working on a few personal deep learning projects with Keras and TensorFlow. However, training models for deep learning with cloud services such as Amazon EC2 and Google Compute Engine isn't free, and as someone who is currently unemployed, I have to keep an eye on extraneous spending and be as cost-efficient as possible (please support my work on Patreon!). I tried deep learning on the cheaper CPU instances instead of GPU instances to save money, and to my surprise, my model training was only slightly slower. As a result, I took a deeper look at the pricing mechanisms of these two types of instances to see if CPUs are more useful for my needs. The pricing of GPU instances on Google Compute Engine starts at $0.745/hr (by attaching a $0.700/hr